Festival of Data Technology

Prevent mis-aligned environments between developers, onboard new-joiners faster, and reduce the time it takes to take your project to production. Sounds interesting? Devcontainers can help you with this. Devcontainers allow you to connect your IDE to a running Docker container and develop inside it. This gives you all the benefits of reproducibility that Docker is known for. In this talk, I will walk you through what Devcontainers are, why they might be useful for you, and how to create one for your Python project using VSCode.

Deploy your own modern data stack using open source components usingTerraform cloud-agnostic tooling. By leveraging open-source components you can deploy a state-of-the-art modern data platform in a day. What are the pro's and con's of “build-it-yourself" in the data+analytics space?

Time-series data is all around us: from logistics to digital marketing, from pricing to stock markets - it’s hard to imagine a modern business that has no time series data to forecast. However, mastering such forecasting is not an easy task. For this talk, we have collected a list of common time series issues that digital fortune tellers commonly run into. You will learn how to identify, understand and resolve them better. This will include stabilising divergent time series, handling outliers without anomaly propagation, reducing the impact of noise and more.

Many machine learning models we use today have the core assumption that our data needs to be tabular, but how often is this truly the case? What if our data points are not independent? By ignoring the potential interrelatedness of our data, do we lose meaningful information that our models cannot leverage? In this talk, we shall explore graph neural networks and highlight how they can solve interesting problems in a way that is intractable when limiting ourselves to using tabular data. We will look at the limitations of common algorithms and highlight how some clever linear algebra enables us to incorporate more meaningful information into our models. Social network data is a popular example of where relationships are relevant but relationships exist in many types of data where it may not be so obvious. Whether it's e-commerce, logistics or molecular data, relationships within your data likely exist and making use of them can be incredibly powerful. This talk will hopefully spark your curiosity and provide you with a way of looking at problems from a new angle. It is intended for anyone with an interest in machine learning and will only lightly touch on some technical details.

Only 53% of the AI applications make it from a prototype stage to production” (Gartner). Companies often find it hard to create and maintain a stable development and production environment for AI and machine learning. For a robust MLOps process the following items are key: data collection is standardized, model training is automated and versioned, model performance is monitored continuously, and feedback is provided on explainable results with the human-in-the-loop. This talk highlights these items by using our automated visual inspection system as an example which we use to inspection our in-house designed and manufactured electronics.

Data Centric AI is the term coined by AI pioneer Andrew Ng for the movement that argues we shift our focus towards iterating on our data instead of models to improve machine learning predictions. But isn't this what we have always done? Why is this trend relevant now? Has something really changed, and if so, how does that change your work as a data scientist? This talk will feature anecdotes and real-world examples of 'model-itis' that serve as an argument for data-centric AI, our lessons learned from winning the Data Centric AI competition, and practical tips on how you can integrate data-centric principles in your daily work.

Do you know the feeling you see all these awesome new concepts like feature stores and automatic retraining, but your organization still struggles to get a simple Machine Learning model in production? Or, you hear all this buzz around MLOps but don’t know if and how to ride the hype train? Look no further. With our MLOps Serious Game, we aim to demystify the world of MLOps. We’ll show you what all the fuzz is about, we let you experience the problems and challenges our clients face and show you (and potentially your manager) why MLOps is the way forward. Caution: you might even have some fun while playing the game!

During the MLOps CodeBreakfast, we will be giving an introduction to MLOps. After this introduction, we will go into more detail on how to implement and deploy a Machine Learning pipeline on both Azure and AWS.

Why should you consider transitioning your analytics team to analytics engineering? We will present the challenges you should expect, and how to approach them.

You probably have heard the news: dbt is not just SQL and jinja anymore. The pythonistas have been dreaming about this for a while, the time when they could use the awesomeness of dbt with their preferred language. But hey, is everything as awesome as promised? What are the benefits of using python in dbt and what are the downsides? When to use SQL and when to use Python? My colleague Padraic and I will share our thoughts on this topic. See you there!

In this talk, we will compare the most widely used BI tools in the market from the perspective of a mature data organization. The focus of this talk WON’T be on flashy features nor superficial sales talk. We will compare both tools in terms of how well they fit in with DataOps best practices. How do they rank in terms of speed of delivery, governance, robustness, and analytical capabilities.

A panel discussion about the future state of Analytics Engineering. What are the best practices in terms of enabling data teams within the organization? Data democratization, what is it and where to start with democratizing your data? Let's discuss with you, Bram, Guillermo, Juan, and Dumky.